On Mon, Sep 16, 2013 at 09:19:23AM +0200, Gilles Louppe wrote:
> So basically, do we agree that the goal of our proposal to this
> workshop will only be to further promote the project in the scientific
> community?
Sounds good to me :)
> Besides that, I like the idea of showing examples to highli
I looked around, but was unable to find a fuzzy clustering algorithm in
sklearn (something analogous to fcm in Matlab) - I did however find an old
gist https://gist.github.com/mblondel/1451300 . Is this in sklearn already
in some form and I am just missing it?
Kyle
---
Hi list,
Just wondering whether anyone on here in planning on attending EMNLP. I'll
be there, and as a heavy user (and hopeful eventual contributor), I'd love
to meet with some of you.
Fred.
--
October Webinars: Code for
Hi Kyle,
personally, I'd love to see SAX in sklearn or any other python library that
I could easily use with sklearn. We don't have any time-series specific
functionality yet (eg. lagged features transformer). So if we choose to add
time-series functionality we should also consider the basics.
Le
2013/9/25 Gilles Louppe :
> I knew I was an outlier ;)
I think it's learned that tree huggers are orthogonal to text mongers,
with Andreas being, of course, our top-notch NLP guy.
--
October Webinars: Code for Performance
On 25 September 2013 19:05, Andreas Mueller wrote:
> On 09/25/2013 06:44 PM, Olivier Grisel wrote:
>> 2013/9/25 Andreas Mueller :
>>> On 09/25/2013 04:15 PM, Jacob Vanderplas wrote:
Very cool!
One quick comment: I'd probably normalize the values in the sparse
matrix to 1. As it's w
I have recently been working with time-series data extensively and looking
at different ways to model, classify, and predict different types of
time-series.
One algorithm I have been playing with is called SAX (
http://www.cs.ucr.edu/~eamonn/SAX.htm). It is a very straightforward
algorithm (basica
On 09/25/2013 06:44 PM, Olivier Grisel wrote:
> 2013/9/25 Andreas Mueller :
>> On 09/25/2013 04:15 PM, Jacob Vanderplas wrote:
>>> Very cool!
>>> One quick comment: I'd probably normalize the values in the sparse
>>> matrix to 1. As it's written, a user with, say, 1 commit on a file
>>> will be co
On Wed, Sep 25, 2013 at 6:33 PM, Andreas Mueller
wrote:
>
> soo why is there no PCA embedding and Isomap? ;)
>
With gravatars would be nice :)
--
October Webinars: Code for Performance
Free Intel webinars can help you
2013/9/25 Andreas Mueller :
> On 09/25/2013 04:15 PM, Jacob Vanderplas wrote:
>> Very cool!
>> One quick comment: I'd probably normalize the values in the sparse
>> matrix to 1. As it's written, a user with, say, 1 commit on a file
>> will be considered a closer neighbor to a user with 0 commits o
On 09/25/2013 04:15 PM, Jacob Vanderplas wrote:
> Very cool!
> One quick comment: I'd probably normalize the values in the sparse
> matrix to 1. As it's written, a user with, say, 1 commit on a file
> will be considered a closer neighbor to a user with 0 commits on that
> file than to a user wi
We need to do an example with that!
G
--
October Webinars: Code for Performance
Free Intel webinars can help you accelerate application performance.
Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most
Indeed, I think normalization is important, depending on what you want
to show. Feel free to play with this if you have good ideas. This is
merely a quick proof of concept.
Also, I would be curious to apply and visualize our new biclustering
algorithms on this.
On 25 September 2013 16:15, Jacob V
Very cool!
One quick comment: I'd probably normalize the values in the sparse matrix
to 1. As it's written, a user with, say, 1 commit on a file will be
considered a closer neighbor to a user with 0 commits on that file than to
a user with 3 commits on that file.
Jake
On Wed, Sep 25, 2013 at
Hi Ron,
The reason why .predict() and .predict_proba doesn't agree is about the
method (Plott's scaling) by which probability values are generated. You can
have a look at my answer here:
http://stackoverflow.com/questions/17017882/scikit-learn-predict-proba-gives-wrong-answers/17142391#17142391
if
On 25 September 2013 13:55, Olivier Grisel wrote:
> 2013/9/25 Luca Cerone :
> >> > (this is not explained in the user guide
> >> >
> >> >
> http://scikit-learn.org/stable/modules/linear_model.html#logistic-regression
> ,
> >> > though).
> >>
> >> All our classifiers support multiclass classificat
2013/9/25 Lars Buitinck :
> 2013/9/25 Olivier Grisel :
>> I don't remember who first did the upload of the 0.14.1 release but it
>> is indeed very possible that there was a glitch in the release process
>> and the first upload was messed up and that an override was required
>> quickly after the fir
2013/9/25 Luca Cerone :
>> > (this is not explained in the user guide
>> >
>> > http://scikit-learn.org/stable/modules/linear_model.html#logistic-regression,
>> > though).
>>
>> All our classifiers support multiclass classification and this is
>> documented in various places.
>
>
> I am sorry, but
>> There are still a few things that are not clear to me from the
>> documentation. Can you customize the classifier to perform a different
>> decision function?
>
> You can subclass it and override the decision_function method.
While true, this can be misleading. You're just changing the final
st
2013/9/25 Olivier Grisel :
> I don't remember who first did the upload of the 0.14.1 release but it
> is indeed very possible that there was a glitch in the release process
> and the first upload was messed up and that an override was required
> quickly after the first upload.
Wasn't that the Reut
2013/9/25 Luca Cerone :
> I am sorry, but I went into the user documentation for logistic regression
> and multiclass classification and didn't find any information about it
Hm, maybe we should put this in a more prominent place like the
tutorial. I'll check the docs if I have time.
> for the pen
>
> > (this is not explained in the user guide
> >
> http://scikit-learn.org/stable/modules/linear_model.html#logistic-regression
> ,
> > though).
>
> All our classifiers support multiclass classification and this is
> documented in various places.
>
I am sorry, but I went into the user documentat
Hi,
I have just put together a quick and dirty script that does that. It
extracts the number of commits for all developers, for all files on a
git directory. It then computes the 3 nearest neighbors for all
contributors.
See the gist below for code and output.
https://gist.github.com/glouppe/6698
2013/9/25 Luca Cerone :
> This morning I checked the source for LogisticRegression in
> sklearn/linear_model/logistic.py and realized that by default it performs
> multiclass classification
> (this is not explained in the user guide
> http://scikit-learn.org/stable/modules/linear_model.html#logisti
Dear Olivier,
thanks for your reply.
On 25 September 2013 10:39, Olivier Grisel wrote:
> LogisticRegression is a already multiclass classifier by default using
> the One vs Rest / All strategy by default (as implemented internally
> by liblinear which LogisticRegression is a wrapper of). So you
LogisticRegression is a already multiclass classifier by default using
the One vs Rest / All strategy by default (as implemented internally
by liblinear which LogisticRegression is a wrapper of). So you don't
need to use OneVsRest in this case.
If you want more info on multiclass reductions here i
I don't remember who first did the upload of the 0.14.1 release but it
is indeed very possible that there was a glitch in the release process
and the first upload was messed up and that an override was required
quickly after the first upload.
pypi gives me 503 currently so that I cannot download t
27 matches
Mail list logo
